Adam Twardoch -- FontLab 6?

charles ellertson's picture

Adam,

We got a notice that the large discount to purchase the Fontographer upgrade would expire in a few days, at the end of the month. Now, the only thing I would use Fontographer over FontLab for is it's *change weight* feature (as opposed to FL's *bold*). If I do only that in Fontographer, I'll (1) spend an extra $100, and (2) have a cumbersome workflow.

It is rumored that when FL Studio 6 comes out, the "bold" will work much better, akin to Fontographer's "change weight." Does anyone at FontLab have a projection when version 6 will come out?

(Jeez, from your web page, you are partly the marketing manager. Usually I don't ask a marketing manager if I can save some money by not buying one of their products, but in FontLab's case, I figure you guys have our interests at heart.)

tourdeforce's picture

@ Jens

Then that's good to hear. Didn't had a chance to try it in any Windows 64 bit OS yet.

There was no problem at all, I just thought that FLS6 could have some issues with 64 bit OS and wanted to ask Adam if they were thinking about it in advance. That's all. Cause (theoretically) if something is not made to be used in another environment, then there might be a chance that something goes wrong. But it's cool to know that FLS 5.x works fine.

twardoch's picture

Tourdeforce,

Most applications made for Windows are 32-bit. Also many Mac OS X applications are 32-bit (although the OS is natively 64 bit since 10.6). For example, Google Chrome, Mozilla Firefox, Skype, Adobe InDesign CS5, Adobe Illustrator CS5 and many many others. They all work fine on a 64-bit system. As far as I could tell, FontLab Studio 5.1 (Mac) and 5.0.4 (Win) work well there as well.

The only slight complication is when you heavily use Python and C-compiled modules. Some C-compiled modules made for Python may be 64-bit only (though it's still a rarity). Or at the very least you need to be aware of that -- so for example, on Windows 7 64-bit, you need to install the 32-bit version of Python 2.4 rather than the 64-bit version.

Best,
Adam

twardoch's picture

Frank,

Which OS is this? Which FLS? Which rasterizer is selected?

Frode Bo Helland's picture

Mac, Lion, FLS 5.1.1, MS Greyscale rasterizer. And while you’re talking: why didn’t you include the Cleartype rasterizer? Who but me hints for Standard GDI anymore?

twardoch's picture

Frank,

integrating ClearType requires more development work. Given the pressure (release of Mac OS X 10.7 Lion), we decided to postpone that to FontLab Studio 6.

Best,
Adam

Cristobal Henestrosa's picture

Adam said:

> Many people are used to it, they're very proficient in the tool but have been asking for more than just UFO.
> The new product will initially be more in the same realm as Glyphs or RoboFab, I suppose. Only with time, it (or perhaps a separate tool) may also cover other parts which are currently unique to FLS. ¶ With UFO, fortunately, we can have interoperability between all these products.

Adam, I don’t know if it’s a good idea to ask this at this moment, but I am trying to figure out if UFO native support is planned for FLS6 or if it will be available only for “FLS7”.

dezcom's picture

Clearly, ufo support must be part of FLS-6. Interoperability between all players in the industry is a must.

I assume this would also affect the way FLS deals with tables. Mostly, the way class key glyphs are notated. Instead of requiring the foot marker as an indicator, why not just the first glyph in the list in the class?

twardoch's picture

Cristobal,

as per Tal's recommendation, we definitely want to have UFO2 support in FontLab Studio 6. "FLS7" will likely support both UFO2 and UFO3.

Best,
Adam

oribendor's picture

Adam,

Why is it a limitation for FontLab to support only integer coordinates? Is there any font format that supports non-integer coordinates? And if not, why should a font editor do?

oribendor's picture

Adam,

Why is it a limitation for FontLab to support only integer coordinates? Is there any font format that supports non-integer coordinates? And if not, why should a font editor do?

oribendor's picture

Adam,

Why is it a limitation for FontLab to support only integer coordinates? Is there any font format that supports non-integer coordinates? And if not, why should a font editor do?

hrant's picture

As many Fontographer old-timers will tell you, using fractional
coordinates will only cause trouble. It gives you an illusion of
accuracy while destroying actual accuracy. Just increase the EM.

hhp

twardoch's picture

Ori,

theoretically, OpenType PS (.otf) supports fractional coordinates, but there are some technical caveats associated with it. In principle, we could say that the final font formats only support integer coordinates but during the work process, having fractional coordinates would be helpful.

For example, if you have drawn your glyphs on an integer grid, but then you'd like to make some adjustments such as: make it slightly more narrow, then slightly wider, then perhaps slant it by a few degrees and then slant it back, or make some tiny rotations, or maybe scale the glyph down and then up again -- on an integer grid, the result will have accumulated the rounding errors from every one of these operations. On a fractional grid, all your shapes (the positions of the BCPs, the angles, the stem thicknesses etc.) will always remain just as in the original design.

Therefore, a hardwired integer grid such as the one in FontLab Studio does impose a certain limitation onto the type designer. A fractional grid such as the one in Fontographer allows you to avoid those rounding errors. But of course, at the very end, when you decide to generate or ship the font, you'll need to align your points to an integer grid, or the software will do this for you.

Hopefully that helps to answer your question!

Best,
Adam

oribendor's picture

Thanks, Adam, you make a very good point.

(Sorry I've posted my previous comment three times, I didn't realize I was doing that.)

Ramiro Espinoza's picture

Are there any news about FL 6? It is gonna be released in the next 3 or 6 years?

hrant's picture

I don't think ridicule is going to speed things up.

hhp

Lasko Dzurovski's picture

Any news about FL 6?

Thank you

hrant's picture

Not to make light of the indefinite delays, but I thought this was both comical and cogent:
http://fuckyeahtypefacedesign.tumblr.com/post/61863698353/its-not-the-to...

hhp

abattis's picture

Hrant, its gone!

yaneczech's picture

Dear Adam,
I would appreciate some news about the state of development as well... (news about beta testing would sound great) :) I am considering switching to Glyphs, but I am used to FL...
Thank you

Jan Janecek

hrant's picture

Dave, I guess the owner of FYTD is in bed with some/all of the SuperRoboGlyphsPolator people and/or Dunwich Type. Cronyism - yay!

BTW since then there's been another moral failing:
http://fuckyeahtypefacedesign.tumblr.com/post/67465965024/i-put-so-much-...
It turns out James Raftopoulos –the winner of the competition– submitted no such thing. Libel - yay!

After all, what use is ethical behavior if it obstructs a cheap laugh, right?

--

And what Jan said.

hhp

Sindre's picture

I have given up waiting, and now use RoboFont on a Mac. I really, really, really wanted to go on using an updated FontLab (and by no means am I an Apple fan boy), but I also need un-rusty, modern tools. (I guess FontLab 6 now will be released in a month or so, given my usual luck.)

Sindre's picture

It is so very sad that two of the things I really loved the most about type and type drawing, FontLab and Typophile, are both dying (or so it seems).

dezcom's picture

Sindre,
I hope you are finding other things to fill the void you feel.

Sindre's picture

Yes, I've taken to the bottle.

dezcom's picture

Then, Salud! or perhaps Skaal?

hrant's picture

Sindre, I think Typophile has picked up. FontLab: I remain hopeful; the fact that they're still sponsoring conferences is a good sign.

hhp

Sindre's picture

Skål! (There is only Darjeeling in my cup, but it does have a noticeable muscatel flavour, so I can pretend it is a Sauternes or something.)

I wonder if type designers at large have abandoned FontLab now for lack of updates, by the way.

Sindre's picture

I remain hopeful too, Hrant, and will definitely upgrade if/when it comes out. A UFO-supporting FL6 could easily be integrated into a RoboFont-based workflow, as I really want the best from both worlds. I am so used to the very good drawing tools of FontLab that I still prefer them.

Thomas Phinney's picture

FontLab has been re-architecting themselves on a new codebase. TransType 4 is the first app to come out using the new codebase. I imagine we will hear more soon.

dezcom's picture

soon?

David Somers's picture

It would be great if FL6 was out this year... but their "re-architecting" seems to be taking a very very very long time (just look how old this thread is!)... TransType 4, which uses the new codebase did appear and does show some promise (its blazingly fast)... but I'm highly confused by the recent release of FontLab Pad because I'd rather see FL6 first. So, perhaps something soon... perhaps... maybe... pretty please?

Thomas Phinney's picture

What I like most about TransType 4 is not just the speed, but the really first-rate UI. :)

I've been doing a little work for the FontLab folks, but I can’t reveal product plans. I will just say that I have reviewed a lot of specs.

hrant's picture

Good news, Thomas.

hhp

John Hudson's picture

Every time I see Adam, which is usually at least once a year, I ask for a progress report. The good news is that there is always progress, and some of it is very cool. The bad news, from my perspective, is that updating to FontLab 6 might require as much of a re-investment in Python scripting as switching to RoboFont or Glyphs.

This is mostly guesswork, but I think FontLab are a little wrong-footed by the recent advent of competition in the primary glyph tool category from RoboFont and Glyphs, just as they were previously wrong-footed by the initial poor implementation of the Mac version of FontLab Studio, which contributed to the development of that competition. It took a lot of effort from what I understand to be a reduced team to remake a stable version of FLS for Mac, and that is part of the reason why 'Victoria' -- the new code architecture for FontLab 6 -- has been delayed. Remember, FontLab has always been primarily a Windows tool and only secondarily ported to Mac. I was reflecting recently that it was about 20 years ago that I saw an ad in a desktop publishing magazine and bought my first license for FontLab 2.5. It was good, and successive (Windows) versions of FontLab have remained good. Current trends favour more openly scriptable object structures, which is a good thing, but it terms of core functionality FLS remains better than any other tool I have tried, and even with its current scripting model fits nicely into my sfnt-centric workflow. I also remain impressed by the size and performance of the VFB source format, even for large fonts of the kinds I regularly build, compared to UFO, which contains less information in much larger files and takes ages to load even in dedicated UFO tools. While I can understand why people are keen for a version of FontLab that natively handles UFO files, I'd like to see the VFB format open-sourced and supported in UFO tools.

Rob O. Font's picture

"...a little wrong-footed..."

A kind of Hobbit. But if you call them that, they shout, "We're Little Wrong-Feets!"

Current trends favour more openly scriptable object structures, but only if extendable. Is that VFB or sfnt?

Time management concerned with figurative "ages to load", while waiting literally "years to do", seem reasonable?

dezcom's picture

I do not notice any difference in loading time between UFO and VFB files on my 3 year old Mac.

The only time I notice quickly is when there is a software or bug issue that prevents me from working well and efficiently. Whatever FontLab does to make Victoria dance needs to be rock solid above all. I much prefer rock solid to glitzy and cute features. The main feature that needs attention is search and replace.

I have licensed and support the further development of the 3 major players, FontLab, Robofont, and Glyphs and hope they all prosper and have a lovely synergistic competition among themselves. Competition is vital. We can never trust putting all of our eggs in one basket again. I have seen this in Quark, Adobe, FontLab, Apple, and Microsoft. When a product becomes the default standard, developers get complacent and the user gets screwed. May healthy competition thrive and keep all parties active and honest.

hrant's picture

But the hobbits did save the day in the end.

hhp

twardoch's picture

We’ve been releasing the things first which were ready first.

TransType 4 was primarily made so that we could test the fully rewritten and modernized import/export infrastructure (OTF, TTF, EOT, WOFF, SVG, VFB, UFO and all the other formats, including conversions between them). We basically wanted the “gear” done and stable — before releasing a follow-up to FontLab Studio. We figured, it’s more important to have a tool with limited functionality which generates good fonts than to have a tool with full functionality that generates bad fonts.

FontLab Pad was done because it was very easy — it’s essentially the Preview panel of TransType. In the same vein, we wanted to have a tool for ourselves — a testing app for the color fonts in all the relevant formats (sbix, COLR, SVG).

We’re heavily under way on glyph editing right now. The lack of my activity here on Typophile and other forums actually indicates that we’re working hard.

We’ll start publishing pre-release versions of our new font editor in the next months.

You’ll love it, and you’ll be blown away :)

A.

hrant's picture

I love it when loyalty pays off. Hopefully soon!

hhp

oldnick's picture

I recently purchased TransType 4 because i had a large number of fonts to generate. Overall, I have been very pleased with the results, except...

when TTF fonts are run through Font Validator, these error messages appear 100% of the time...

W0020—Tables are not in optimal order: table 'glyf' precedes table 'loca'

OS/2
E2132—The version number is invalid: 4
E2127—The table length does not match the expected length for this version
W0051—Cannot perform test due to other errors in this table: OS/2 table appears to be corrupt. No further tests will be performed.

gasp
E1003—The version number is not 0: 1
E1000—Some rangeGaspBehavior fields contain invalid flags

bojev's picture

Only used TransType a couple of times since I got the update but like others it is very zippy and seems to work well.

twardoch's picture

Nick,

Microsoft Font Validator has not been updated in a long time. The OS/2 table version 4 and the gasp table version 1 have been added to the OpenType standard after the release of Font Validator, so the complaints related to OS/2 and gasp are all a false alarm. The tables are correct.

As for the "W0020—Tables are not in optimal order: table 'glyf' precedes table 'loca'" warning, it's by no means critical but we'll look into it.

Thanks,
Adam

John Hudson's picture

Nick, with regard to the table order warning, this is a test that Microsoft put into Font Validator based on performance testing that they did in the 1990s. If the tables are in a particular order, Windows can do some things faster. If you're concerned about this, you can run the FastFont command line tool from MS on your fonts, which will put the tables in the 'optimal' order.

Adam, I don't know if the preferred table order is published anywhere, but you presumably run FastFont on a font and then look what the resulting table order is. If your new tools would write that order by default, it would save us a post-productions step.

k.l.'s picture

It is mentioned in the OT spec's recommendations: https://www.microsoft.com/typography/otspec/recom.htm Search for section 'Optimized Table Ordering'.

oldnick's picture

John,

I routinely use FastFont to order the tables correctly, and it's a breeze: simply drop the font filename onto fastfont.exe, and it does the job in the blink of an eye. During my latest rounds of updating, I had to do this over 600 times, which became a bit tedious. I pointed out the anomaly in order to to save others the effort, because it seems to me that the fix should be easy.

blokland's picture

Nick: ‘[…] I had to do this over 600 times, which became a bit tedious.

TBH, I never used FastFont, but just had a look at it. Actually you can make a shell script for FastFont (using a listing of all your .ttf files). This is a bit less tedious ;-)

FEB

Sindre's picture

Great, great news. So looking forward.

Nick Cooke's picture

I'm looking forward to the new FL release. Working on my latest typeface I have a set of swashes. I have been laboriously copying from 'results' in the Open Type Features Preview panel, then pasting into the Metrics panel to do the class kerning. It would be great if any features results could be incorporated into the main metrics panel for ease of editing. That would be a huge improvement.

Syndicate content Syndicate content